ETRI Journal

Electronics and Telecommunications Research Institute (한국전자통신연구원)
권호 표지
DOI QR코드

DOI QR Code

Amazon product recommendation system based on a modified convolutional neural network

  • Yarasu Madhavi Latha (Department of Computer Science and Engineering, Koneru Lakshmaiah Educational Foundation) ;
  • B. Srinivasa Rao (Department of Computer Science and Engineering, Koneru Lakshmaiah Educational Foundation)
  • Received : 2023.04.25
  • Accepted : 2023.11.29
  • Published : 2024.08.20
Download PDF
⟨ Previous Next ⟩

Abstract

In e-commerce platforms, sentiment analysis on an enormous number of user reviews efficiently enhances user satisfaction. In this article, an automated product recommendation system is developed based on machine and deep-learning models. In the initial step, the text data are acquired from the Amazon Product Reviews dataset, which includes 60 000 customer reviews with 14 806 neutral reviews, 19 567 negative reviews, and 25 627 positive reviews. Further, the text data denoising is carried out using techniques such as stop word removal, stemming, segregation, lemmatization, and tokenization. Removing stop-words (duplicate and inconsistent text) and other denoising techniques improves the classification performance and decreases the training time of the model. Next, vectorization is accomplished utilizing the term frequency-inverse document frequency technique, which converts denoised text to numerical vectors for faster code execution. The obtained feature vectors are given to the modified convolutional neural network model for sentiment analysis on e-commerce platforms. The empirical result shows that the proposed model obtained a mean accuracy of 97.40% on the APR dataset.

Keywords

References

  1. R. Liang and J. Wang, A linguistic intuitionistic cloud decision support model with sentiment analysis for product selection in e-commerce, Int. J. Fuzzy Syst. 21 (2019), no. 3, 963-977. DOI 10.1007/s40815-019-00606-0.
  2. Q. Sun, J. Niu, Z. Yao, and H. Yan, Exploring eWOM in online customer reviews: sentiment analysis at a fine-grained level, Eng. Appl. Artif. Intel. 81 (2019), 68-78. DOI 10.1016/j.engappai.2019.02.004.
  3. B. Ozyurt and M. A. Akcayol, A new topic modeling based approach for aspect extraction in aspect based sentiment analysis: SS-LDA, Expert Syst. Appl. 168 (2021), 114231. DOI 10.1016/j.eswa.2020.114231.
  4. K. Wang, T. Zhang, T. Xue, Y. Lu, and S.-G. Na, E-commerce personalized recommendation analysis by deeply-learned clustering, J. Visual Commun. Image Represent. 71 (2020), 102735. DOI 10.1016/j.jvcir.2019.102735.
  5. A. Iftikhar, M. A. Ghazanfar, M. Ayub, Z. Mehmood, and M. Maqsood, An improved product recommendation method for collaborative filtering, IEEE Access 8 (2020), 123841-123857. DOI 10.1109/ACCESS.2020.3005953.
  6. S. G. K. Patro, B. K. Mishra, S. K. Panda, R. Kumar, H. V. Long, D. Taniar, and I. Priyadarshini, A hybrid action-related K-nearest neighbour (HAR-KNN) approach for recommendation systems, IEEE Access 8 (2020), 90978-90991. DOI 10.1109/ACCESS.2020.2994056.
  7. M. Shaheen, S. M. Awan, N. Hussain, and Z. A. Gondal, Sentiment analysis on mobile phone reviews using supervised learning techniques, Int. J. Mod. Educ. Comput. Sci. 11 (2019), no. 7, 32-43. DOI 10.5815/ijmecs.2019.07.04.
  8. T. U. Haque, N. N. Saber, and F. M. Shah, Sentiment analysis on large scale Amazon product reviews, In Proc. IEEE International Conference on Innovative Research and Development [ICIRD], Bangkok, Thailand, 2018, 1-6.
  9. S. Wassan, X. Chen, T. Shen, M. Waqar, and N. Z. Jhanjhi, Amazon product sentiment analysis using machine learning techniques, Rev. Argentina Clin. Psicologica 30 (2021), no. 1, 695. DOI 10.24205/03276716.2020.2065.
  10. F. Zhang, X. Hao, J. Chao, and S. Yuan, Label propagationbased approach for detecting review spammer groups on e-commerce websites, Knowl. Based Syst. 193 (2020), 105520. DOI 10.1016/j.knosys.2020.105520.
  11. A. S. Ghabayen and B. H. Ahmed, Polarity analysis of customer reviews based on part-of-speech subcategory, J. Intell. Syst. 29 (2020), no. 1, 1535-1544. DOI 10.1515/jisys-2018-0356.
  12. F. Xu, Z. Pan, and R. Xia, E-commerce product review sentiment classification based on a naive Bayes continuous learning framework, Inf. Process. Manage. 57 (2020), no. 5, 102221. DOI 10.1016/j.ipm.2020.102221.
  13. L. Yang, Y. Li, J. Wang, and R. S. Sherratt, Sentiment analysis for E-commerce product reviews in Chinese based on sentiment lexicon and deep learning, IEEE Access 8 (2020), 23522-23530. DOI 10.1109/ACCESS.2020.2969854.
  14. A. Onan, Sentiment analysis on product reviews based on weighted word embeddings and deep neural networks, Concurrency Comput. Pract. Exper. 33 (2021), no. 23, e5909. DOI 10.1002/cpe.5909.
  15. A. Suresh and M. J. C. M. Belinda, Online product recommendation system using gated recurrent unit with Broyden fletcher Goldfarb Shanno algorithm, Evol. Intell. 15 (2022), no. 3, 1861-1874. DOI 10.1007/s12065-021-00594-x.
  16. L. Kolhe, A. K. Jetawat, and V. Khairnar, Robust product recommendation system using modified grey wolf optimizer and quantum inspired possibilistic fuzzy C-means, Cluster Comput. 24 (2021), no. 2, 953-968. DOI 10.1007/s10586-020-03171-6.
  17. J. Shobana and M. Murali, An efficient sentiment analysis methodology based on long short-term memory networks, Complex Intell. Syst. 7 (2021), no. 5, 2485-2501. DOI 10.1007/s40747-021-00436-4.
  18. Y. Zhang, Z. Liu, and C. Sang, Unifying paragraph embeddings and neural collaborative filtering for hybrid recommendation, Appl. Soft Comput. 106 (2021), 107345. DOI 10.1016/j.asoc.2021.107345.
  19. B. M. Shoja and N. Tabrizi, Customer reviews analysis with deep neural networks for e-commerce recommender systems, IEEE Access 7 (2019), 119121-119130. DOI 10.1109/ACCESS.2019.2937518.
  20. T. Mandhula, S. Pabboju, and N. Gugulotu, Predicting the customer's opinion on Amazon products using selective memory architecture-based convolutional neural network, J. Supercomput. 76 (2020), no. 8, 5923-5947. DOI 10.1007/s11227-019-03081-4.
  21. P. Hajek, A. Barushka, and M. Munk, Fake consumer review detection using deep neural networks integrating word embeddings and emotion mining, Neural Comput. Appl. 32 (2020), no. 23, 17259-17274. DOI 10.1007/s00521-020-04757-2.
  22. M. Umer, Z. Imtiaz, M. Ahmad, M. Nappi, C. Medaglia, G. S. Choi, and A. Mehmood, Impact of convolutional neural network and FastText embedding on text classification, Multimedia Tools Appl. 82 (2023), no. 4, pp. 5569-5585. DOI 10.1007/s11042-022-13459-x.
  23. S. Sachin, A. Tripathi, N. Mahajan, S. Aggarwal, and P. Nagrath, Sentiment analysis using gated recurrent neural networks. SN Comput. Sci. 1 (2020), no. 2, 74. DOI 10.1007/s42979-020-0076-y.
  24. A. Alsayat, Improving sentiment analysis for social media applications using an ensemble deep learning language model, Arab. J. Sci. Eng. 47 (2022), no. 2, 2499-2511. DOI 10.1007/s13369-021-06227-w.
  25. N. Ghasemi and S. Momtazi, Neural text similarity of user reviews for improving collaborative filtering recommender systems, Electron. Commer. Res. Appl. 45 (2021), 101019. DOI 10.1016/j.elerap.2020.101019.
  26. A. Onan, Bidirectional convolutional recurrent neural network architecture with group-wise enhancement mechanism for text sentiment classification, J. King Saud Univ. - Comput. Inform. Sci. 34 (2022), no. 5, 2098-2117.
  27. A. Onan, Consensus clustering-based undersampling approach to imbalanced learning, Sci. Program. 2019 (2019), 1-14.
  28. A. Onan, S. Korukoglu, and H. Bulut, Ensemble of keyword extraction methods and classifiers in text classification, Expert Syst. Appl. 57 (2016), 232-247.
  29. A. Onan, Two-stage topic extraction model for bibliometric data analysis based on word embeddings and clustering, IEEE Access 7 (2019), 145614-145633.
  30. A. Onan and S. Korukoglu, A feature selection model based on genetic rank aggregation for text sentiment classification, J. Inf. Sci. 43 (2017), no. 1, 25-38.
  31. A. Onan, S. Korukoglu, and H. Bulut, A hybrid ensemble pruning approach based on consensus clustering and multi-objective evolutionary algorithm for sentiment classification, Inf. Process. Manage. 53 (2017), no. 4, 814-833.
  32. A. Onan, Mining opinions from instructor evaluation reviews: a deep learning approach, Comput. Appl. Eng. Educ. 28 (2020), no. 1, 117-138.
  33. A. Onan, Sentiment analysis on massive open online course evaluations: a text mining and deep learning approach, Comput. Appl. Eng. Educ. 29 (2021), no. 3, 572-589.
  34. A. Onan, An ensemble scheme based on language function analysis and feature engineering for text genre classification, J. Inf. Sci. 44 (2018), no. 1, 28-47.
  35. A. Onan and M. A. Tocoglu, A term weighted neural language model and stacked bidirectional LSTM based framework for sarcasm identification, IEEE Access 9 (2021), 7701-7722.
  36. A. Onan, Topic-enriched word embeddings for sarcasm identification, in Software Engineering Methods in Intelligent Algorithms, In Proceedings of 8th Computer Science On-line Conference 2019, Vol. 18, Springer International Publishing, 2019, 293-304.
  37. A. Onan, Biomedical text categorization based on ensemble pruning and optimized topic modelling, Comput. Math. Methods Med. 2018 (2018), 2497471.
  38. A. Onan, H. Bulut, and S. Korukoglu, An improved ant algorithm with LDA-based representation for text document clustering, J. Inf. Sci. 43 (2017), no. 2, 275-292.
  39. A. Onan, Hierarchical graph-based text classification framework with contextual node embedding and BERT-based dynamic fusion, J. King Saud Univ. - Comput. Inform. Sci. 35 (2023), 101610.
  40. A. Onan, GTR-GA: harnessing the power of graph-based neural networks and genetic algorithms for text augmentation, Expert Syst. Appl. (2023), 120908.
  41. A. Onan, SRL-ACO: a text augmentation framework based on semantic role labeling and ant colony optimization, J. King Saud Univ. - Comput. Inform. Sci. 35 (2023), 101611.
  42. H. Zhao, Z. Liu, X. Yao, and Q. Yang, A machine learning-based sentiment analysis of online product reviews with a novel term weighting and feature selection approach, Inf. Process. Manage. 58 (2021), no. 5, 102656. DOI 10.1016/j.ipm.2021.102656.
  43. M. A. Fauzi, Word2Vec model for sentiment analysis of product reviews in Indonesian language, Int. J. Electr. Comput. Eng. 9 (2019), no. 1, 525-530. DOI 10.11591/ijece.v9i1.pp525-530.
  44. B. Bansal and S. Srivastava, Hybrid attribute based sentiment classification of online reviews for consumer intelligence. Appl. Intell. 49 (2019), no. 1, 137-149. DOI 10.1007/s10489-018-1299-7.
  45. O. Gokalp, E. Tasci, and A. Ugur, A novel wrapper feature selection algorithm based on iterated greedy metaheuristic for sentiment classification, Expert Syst. Appl. 146 (2020), 113176. DOI 10.1016/j.eswa.2020.113176.
  46. M. Hamdi, Affirmative ant colony optimization based support vector machine for sentiment classification, Electronics 11 (2022), no. 7, 1051. DOI 10.3390/electronics11071051.
  47. J. Woo and M. Mishra, Predicting the ratings of Amazon products using big data, Wiley Interdiscip. Rev.: Data min. Knowl. Discov. 11 (2021), no. 3, e1400. DOI 10.1002/widm.1400.
  48. S. Dey, S. Wasif, D. S. Tonmoy, S. Sultana, J. Sarkar, and M. Dey, A comparative study of support vector machine and naive Bayes classifier for sentiment analysis on Amazon product reviews, In 2020 International Conference on Contemporary Computing and Applications [IC3A], Lucknow, India, 2020, 217-220.
  49. T. Bezdan, C. Stoean, A. A. Naamany, N. Bacanin, T. A. Rashid, M. Zivkovic, and K. Venkatachalam, Hybrid fruit-fly optimization algorithm with k-means for text document clustering, Mathematics 9 (2021), no. 16, 1929. DOI 10.3390/math9161929.
  50. S. W. Kim and J. M. Gil, Research paper classification systems based on TF-IDF and LDA schemes. Hum.-Centric Comput. Inform. Sci. 9 (2019), 30. DOI 10.1186/s13673-019-0192-7.
  51. Z. Zhu, J. Liang, D. Li, H. Yu, and G. Liu, Hot topic detection based on a refined TF-IDF algorithm, IEEE Access 7 (2019), 26996-27007. DOI 10.1109/ACCESS.2019.2893980.
  52. S. Tao, C. Shen, L. Zhu, and T. Dai, SVD-CNN: a convolutional neural network model with orthogonal constraints based on SVD for context-aware citation recommendation, Comput. Intell. Neurosci. 2020 (2020), 5343214. DOI 10.1155/2020/5343214.
  53. D. Liu, X. Lai, Z. Xiao, D. Liu, X. Hu, and P. Zhang, Fault diagnosis of rotating machinery based on convolutional neural network and singular value decomposition, Shock Vib. 2020 (2020), 6542913. DOI 10.1155/2020/6542913.